Generative AI Engineer

Generative AI Engineer About the Role

Experience:

5+ years Tech Focus:

Python, RAG, LangChain/LangGraph, OpenAI/Bedrock, SQL/Spark, Postgres/MongoDB, Pandas/scikit-learn/CV2 Cloud:

AWS, GCP, Azure

We’re looking for an AI Engineer with strong Python expertise and hands-on experience delivering systems powered by Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG). You’ll design and ship resilient AI services, build robust orchestration with LangChain/LangGraph, and ensure reliability at scale with strong error handling and

production failovers

. The ideal candidate is a pragmatic builder who can collaborate across engineering, data, and product to ship business impact quickly and safely.

What You’ll Do

Design & Ship LLM/RAG Services:

Build Python-based services and APIs that leverage RAG pipelines (chunking, indexing, embeddings, retrieval) and LLM orchestration with

LangChain

/

LangGraph

. Model Selection & Routing:

Implement provider-agnostic strategies across

OpenAI

,

AWS Bedrock

, and other endpoints; add smart routing, fallback, and

production failover

patterns for high availability. Data & Feature Pipelines:

Develop ingestion, preprocessing, and transformation pipelines with

Python, SQL, Spark

; persist and retrieve from

Postgres

and

MongoDB

. Document & Image Processing:

Use

Pandas

,

scikit-learn

, and

CV2

for unstructured data preparation (e.g., OCR-ready cleaning, page-level segmentation, image enhancements). Reliability & Observability:

Implement robust

error handling

, logging, metrics, tracing, and alerting; design for graceful degradation and circuit breakers. Security & Compliance:

Enforce secrets management, RBAC, data access controls, PII handling, and auditability across services. DevEx & Reuse:

Contribute to reusable libraries, templates, and documentation to accelerate delivery across teams. Cross-Functional Delivery:

Partner with data scientists, platform engineers, and product to scope requirements and ship end‑to‑end solutions.

Required Qualifications

5+ years

of professional software engineering with a track record of independently designing and shipping complex systems. Advanced

Python

with strong software engineering practices (testing, CI/CD, containerization). Practical experience building systems with

LLMs

and

RAG

patterns (prompting, retrieval pipelines, indexing, chunking, embeddings). Orchestration with

LangChain

and

LangGraph

for multi-step workflows. Hands-on with

OpenAI

and

AWS Bedrock

(usage, quotas, rate limits, provider selection). Data engineering depth with

SQL

,

Spark

, and storage systems (

Postgres

,

MongoDB

). Proficiency with

Pandas

,

scikit-learn

, and

CV2

for data preparation and feature engineering on text/images. Strong

error handling

and resiliency design; proven experience implementing

production failovers

. Cloud proficiency in

AWS

,

GCP

, and/or

Azure

; experience with infrastructure-as-code and container orchestration (e.g., Docker, Kubernetes). Clear communication and ability to collaborate with cross-functional stakeholders.

Core Technology Stack

Languages & Data:

Python, SQL, Spark; Postgres, MongoDB Libraries:

Pandas, scikit-learn,

CV2 LLM & Orchestration:

LangChain

,

LangGraph

,

RAG Model Providers:

OpenAI

,

AWS Bedrock

(plus optional cloud-native endpoints) Cloud:

AWS

,

GCP

,

Azure Reliability:

Model selection/routing,

error handling

,

production failovers

, observability

If interested, please share your resume at:

manisha.2190498@infosys.com Location - PAN India